Histological differences related to autophagy in the minor salivary gland between primary and secondary types of Sjögren's syndrome

Abstract

Some forms of Sjögren's syndrome (SS) follow a clinical course accompanied by systemic symptoms caused by lymphocyte infiltration and proliferation in the liver, kidneys, and other organs. To better understand the clinical outcomes of SS, here we used minor salivary gland tissues from patients and examine their molecular, biological, and pathological characteristics. A retrospective study was performed, combining clinical data and formalin-fixed paraffin-embedded (FFPE) samples from female patients over 60 years of age who underwent biopsies at Okayama University Hospital. We employed direct digital RNA counting with nCounter^® and multiplex immunofluorescence analysis with a PhenoCycler™ on the labial gland biopsies. We compared FFPE samples from SS patients who presented with other connective tissue diseases (secondary SS) with those from stable SS patients with symptoms restricted to the exocrine glands (primary SS). Secondary SS tissues showed enhanced epithelial damage and lymphocytic infiltration accompanied by elevated expression of autophagy marker genes in the immune cells of the labial glands. The close intercellular distance between helper T cells and B cells positive for autophagy-associated molecules suggests accelerated autophagy in these lymphocytes and potential B cell activation by helper T cells. These findings indicate that examination of FFPE samples from labial gland biopsies can be an effective tool for evaluating molecular histological differences between secondary and primary SS through multiplexed analysis of gene expression and tissue imaging.

Keywords: Autoimmune disease; Autophagy; Multiplex immunostaining; Spatial analysis; Xerostomia.

Fig. 1

Comparison of immune-related gene expressions between mucocele and SS by digital RNA counting and RT-qPCR. (A) Volcano plot comparing gene expression between the mucocele and primary SS by nCounter^® system analysis. The left side shows the group of genes for which expression was upregulated in mucoceles. The right side shows the group of genes for which expression was upregulated in primary SS. (B) Volcano plot comparing gene expression between the mucocele and secondary SS. The left side shows the group of genes for which expression was upregulated in mucocele. The right side shows the group of genes for which expression was upregulated in secondary SS. Genes with P-value < 0.5 are highlighted in dark blue. (C) Volcano plot comparing gene expression between primary SS and secondary SS. The left side shows the group of genes for which expression was upregulated in primary SS. The right side shows the group of genes for which expression was upregulated in secondary SS. (D) The percentages of lymphocytic foci areas in the labial gland tissue determined by hematoxylin and eosin (H&E) staining. The Bonferroni method was employed for comparison between the three groups. (E) Volcano plot comparing gene expression between primary SS and secondary SS with highlighting autophagy-related factors. The expression of autophagy-related ten genes (colored) was upregulated in secondary SS. (F) Comparison of Autophagy related 5 (ATG5), LC3B and sequestosome 1 (SQSTM1) expression between primary SS and secondary SS using qRT-PCR. Bars indicate mean values and dots indicate their respective values. ** p < 0.01

Fig. 2

Cell profiling by multiplex immunostaining. (A) Multiplex immunostaining and the corresponding histological (H&E) images are presented. Multiplex immunostaining is 19-color superimposed image by PhenoCycler™ analysis. The three white squares delineate, from left to right, the stroma area, denoted as stroma, the epithelial area, denoted as epithelium, and the lymphocytic foci, denoted as “Immune”, which is the same as immune area in the text. Since epithelial and stromal tissues were intermingled, the panels for “Epithelium” are images rich in epithelial tissues and those for “Stroma” are rich in connective tissue. Scale bars representing 200 μm. (B) H&E image, accompanied by scale bar: 200 μm. (C–Q) Single-cell profiles based on multiplex immunostaining. The top sections show areas of lymphocytic foci (C–G), the middle sections show the epithelium (H–L). and the bottoms sections show the stroma (M–Q). (C, H, M) Localization of E-cadherin (E-cad), aquaporin5 (AQP5), and cytokeratin7 (KRT7) in epithelial tissue. Similarly, the followings are shown: immune cells defined by CD11c, CD3e, and CD8 (D, I, N); vascular lymphatics defined by podoplanin and CD31 (E, J, O); autophagy-related protein markers defined by light chain 3 alpha (LC3B) and SQSTM1 (F, K, P); and the area reactive to the autoantibody SS-A (G, L, R)

Fig. 3

Difference in tissue damage between primary SS and secondary SS. (A, B, E, H) Histological analyses of primary SS and (C, D, F, I) secondary SS samples were performed. Tissue samples were imaged at low magnification (A–D) along with the corresponding H&E images. Scale bar: 200 μm (A–D), 40 μm (E, F, H, I). (E, F) Four-color multiplex immunostaining showing epithelial tissue damage: KRT7, a duct marker; AQP5, an acinar marker; and E-cad, an adhesion factor in epithelial tissues. Localization of CD3e indicates immune cells in the vicinity. White arrowhead in (F) indicates one of the representative cells double positive for KRT7 and AQP5. (G) Difference in the percentage of KRT7-positive cells among AQP5-positive cells. *p < 0.05. (H, I) Four-color multiplex immunostaining of the vascular lymphoid tissue is shown: CD31, a vascular endothelial cell marker; podoplanin, a lymphatic endothelial cell marker; and E-cad, an adhesion factor for epithelial tissue. Localization of CD3e indicates immune cells in the vicinity. White arrowhead in (I) indicates one of the representative CD31-positive cells surrounding the acinar cells. (J) Differences in the percentages of CD31-positive and podoplanin-positive cells relative to the total cell count are shown. *p < 0.05

Fig. 4

Autophagy marker expression in the area of lymphocytic foci. (A) Multiplex immunostaining image of a ten-color overlay of primary SS samples. (B) Corresponding cell-type map from the multiplex immunostaining image using the QuPath spatial analysis toolbox. The lymphocyte infiltration area is circled with a dotted line. (C) Multiplex immunostaining image of LC3B and SQSTM1. The lymphocyte infiltration area is circled with a dotted line. Scale bar: 100 μm. (D) Corresponding cell-type map of the autophagy markers LC3B and SQSTM1. The thin white line indicates the boundary between cells. (E) Correlation between the intensity of LC3B and SQSTM1 protein expression in lymphocyte infiltration areas. (F) Comparison of the expression levels of autophagy markers in primary SS and secondary SS, presented as a relative comparison with primary SS. **p < 0.01

Fig. 5

Comparison of distances between T cells and B cells via cellular neighborhood analysis. (A, B) Multiplex immunostaining images of ten-color overlay images, with the white-circled area in the lymphocytic foci area for analysis. Scale bar: 100 μm. (C, D) Four distinct cellular neighborhoods (CNs) were identified in SS samples: epithelial, B cell-rich, T cell-rich, and mixed. (E, F) The heatmap for marker expression in each area. (G, H) Schematic diagrams comparing the distances between the areas, with the distance between the four areas representing the length of each CN edge distance tangential to each other. The intercellular distance between the T and B cells was measured as the distance between the nuclei. (I) Interaction of T cells and B cells in the secondary SS was significantly closer than that in the primary SS. **p < 0.01. (J) The distances between CD4 + T cells and CD20 + B cells after dividing the intensity of autophagy marker expression between the two groups. *p < 0.05. (K) Difference in INF-γ expression by digital RNA analysis. **p < 0.01